deniskuzZ opened a new pull request, #6922: URL: https://github.com/apache/hive/pull/6922
### What changes were proposed in this pull request? - HiveDeleteFilter applies a data file's PositionDeleteIndex to a whole batch: it marks the deleted rows in the batch's position range and drops them from the selected rows - Removes the per-row path through Iceberg's DeleteFilter (HiveRow, HiveStructLike, HiveValueConverter, the IS_DELETED column) - Equality deletes are still rejected on the vectorized path, as before - Stacked on HIVE-30108 ### Why are the changes needed? Optimization: local 19.0 -> 1.32 ns/row. 1TRC part2 V2, 40 x r5d (with HIVE-30108): delete overhead 179.0 -> 29.4 s, part2 394.3 -> 263.1 s ### Does this PR introduce _any_ user-facing change? No ### How was this patch tested? TestHiveDeleteFilter, TestHiveIcebergVectorization -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
